184 A. Armstrong Agriculture, Ecosystems and Environment 77 2000 179–192
Fig. 2. Comparison of observed and modelled water tables, solid line for four of the monitored fields at Southlake Moor in the Somerset levels and Moors ESA, SW England, 1988–1994.
5. Estimation of soil surface strength
As the ability of birds to probe the soil for food is a major factor in the choice of sites for breeding, the
estimation of soil surface strength is required to iden- tify site suitability for bird feeding needs Tickner and
Evans, 1991. For many soils, there is a relatively well defined relationship between soil strength and water
content. As a first approximation, soil water content at the surface is also correlated with the water table
depth from the surface, so a relationship between soil strength and water table offers the possibility of mod-
elling the suitability of the soil for feeding by birds. This relationship cannot, however, be established from
a priori principles, and must be calibrated from field observation.
The penetration resistance of the soil is a measure of the difficulty that a bird might be expected to have
in feeding. For work specifically related to bird beak penetration, a special penetrometer has been devised
Green, 1986 which records the force required to push a narrow cylinder into the soil a distance of 100 mm,
and so mimics the behaviour of a bird beak. At both sites, the soil surface strength was recorded when-
ever the auger holes were read using this penetrome- ter. On each occasion eight replicate measurements of
soil strength were taken at three locations within each field.
The field data showed very different behaviours for the two sites. At Halvergate, the soil strength mea-
surements Fig. 3, showed no significant correlation between water table depth and soil strength for any of
the six fields. Nearly all the observations are for values greater than a 6 kg force required to penetrate the sur-
face, and this value was maintained even when the soil was flooded. This value is greater than the estimated
maximum force of 3 kg required for penetration of the soil by Snipe Gallinago Gallinago Green, 1988. It
is considered that this high surface penetration resis- tance is the consequence of the grazing management
regime of the marshes, which are subject to short peri- ods of high intensity stocking. The consequence is the
formation of a hard surface mat of vegetation and com- pacted soil. It is suggested that this capping behaviour
is one of the characteristics of clay marshes, such as the Halvergate marshes, compared to peat marshes.
A. Armstrong Agriculture, Ecosystems and Environment 77 2000 179–192 185
Fig. 3. Relationships between soil strength and water table depth for each of the monitored fields in the Halvergate area of the Broads ESA.
The results also serve to emphasise the impact of and agricultural management practice affecting the value
of land for ecological purposes. These results contrast with those obtained at South-
lake Moor, Fig. 4 where a clear correlation between water table depth and soil surface strength has been
observed, for all the fields in the study area. The over- all correlation of 0.69 was considered to be remarkably
good, especially when the other often short term fac- tors that can effect soil penetrability are considered.
For Southlake it was, thus, possible to estimate soil surface strength from the water table position. Linear
regression was used to estimate the relationship be- tween the two variables, and the estimated pattern of
soil surface strength was shown to follow the observed pattern Fig. 5. These data show the pattern of be-
haviour that is expected, being harder in the late sum- mer, coming to their softest in the winter, and gradu-
ally hardening again during the summer, this pattern of behaviour being repeated for all the fields.
6. Estimation of surface flooding
